Atmospheric particulate matter impairs pulmonary barriers by triggering FTH1-mediated ferroptosis

TL;DR

PM2.5 harms lung function by causing oxidative stress and damaging lung barriers through a process called ferroptosis.

Contribution

This study identifies FTH1-mediated ferroptosis as a novel mechanism by which PM2.5 induces lung injury.

Findings

PM2.5 exposure compromises lung function and disrupts tissue structure.

Ferroptosis inhibition with Fer-1 reverses PM2.5-induced damage in lung cells.

FTH1 is a central regulator in PM2.5-induced lung injury.

Abstract

PM2.5 exposure is harmful to health. The related mechanisms by which PM2.5 induced acute lung injury remain to be investigated. Herein, we found PM2.5 compromised lung function, disrupted lung tissue histology, and elevated inflammatory cytokines. Transcriptome sequencing data showed that ferroptosis-mediated oxidative stress plays a critical role in both cellular and murine models. In airway epithelial cells, a dose-dependent decrease in junction proteins was observed following PM2.5 induction, which was subsequently ameliorated by the ferroptosis inhibitor Fer-1 treatment. In alveolar macrophages, continuous exposure to PM2.5 for 6 h resulted in diminished phagocytic capacity, which was also reversed upon the addition of Fer-1. Moreover, network analysis identified FTH1 as a central node in regulating PM2.5-induced lung injury. These findings suggest that enhanced pulmonary uptake and retention of PM2.5 correlate with more severe lung injuries, with the number of barriers encountered by PM2.5 during its transit potentially playing a crucial role. The underlying mechanism is partially through disrupting pulmonary barriers via promoting FTH1-mediated ferroptosis.

•PM2.5 impairs airway epithelial barriers•Barrier injury induced by PM2.5 is mediated by ferroptosis-related oxidative stress•FTH1 mitigates PM2.5-induced barrier damage in lung cells

PM2.5 impairs airway epithelial barriers

Barrier injury induced by PM2.5 is mediated by ferroptosis-related oxidative stress

FTH1 mitigates PM2.5-induced barrier damage in lung cells

Biological sciences; Toxicology; Environmental toxicology

Figures

Acute inflammation in the lungs of mice following three different exposure protocols(A) Schematic overview of the experimental design. Mice were exposed to PM2.5 exposure via three different methods: single intranasal instillation (NE), intratracheal instillation (TE), or whole-body exposure (WE) (blue arrows), followed by a one-day cessation, and this cycle was repeated three times. After the third exposure, all mice were euthanized. Non-invasive pulmonary function tests were conducted after each exposure and cessation period (black arrows).(B) Gross observation of lung tissues.(C) Histologic

Changes of the epithelial barrier in the upper and lower respiratory tracts of mice following various exposure patterns(A) Representative micrographs of nasal and tracheal tissues stained with H&E. Scale bars: 50 μm. Goblet cell metaplasia is denoted by black arrowheads, ciliary structure by green arrowheads, and inflammatory cell influx by red arrowheads.(B) Representative micrographs of nasal and tracheal tissues stained with AB-PAS. Scale bars: upper, 50 μm; below, 100 μm. Nasal mucosubstances are denoted by black arrows, and mucous gland hyperplasia is indicated by orange arrows.(C–F) Repr

PM2.5 impaired the epithelial barrier partially by promoting ferroptosis in BEAS-2B cells(A and B) Western blot analysis and quantitative results of ZO-1, occludin, and E-cad in BEAS-2B cells exposed to varying concentrations of PM2.5 (10, 20, 50, and 100 μg/cm^2^) over different time periods (1 h, 3 h, and 6 h).(C) Volcano plots of differentially expressed genes (DEGs)in BEAS-2B cells from PM2.5 exposure and Control groups in a dataset ([[GSE155616](GSE155616)](GSE155616)). Red points represented upregulated genes and blue points represented downregulated genes filtered based on fold change (

PM2.5 triggered ferroptosis and promoted inflammation in MH-S cells(A) Immunofluorescence analysis and quantification of CD68 expression (white arrows) in the lung tissues of mice subjected to various exposure patterns. Scale bars: 50 μm. Red arrows highlight the particle aggregates.(B) Confocal laser scanning microscopy images illustrating the phagocytic capacity of MH-S following PM2.5 exposure at various time intervals. Yellow arrowheads denote particle aggregates, and white arrowheads indicate filopodia protrusions. Scale bars: 50 μm.(C) The average number of pHrodo green Zymosan biopartic

PM2.5 resulted in ferroptosis by upregulating FTH1 in BEAS-2B and MH-S(A) Hub DEGs were identified from the PPI network using Cytoscape, based on the [[GSE155616](GSE155616)](GSE155616) dataset.(B and C) Western blot analysis of FTH1 in PM2.5-induced BEAS-2B and MH-S with or without Fer-1 addition.(C–E) Flow cytometry results of intracellular ROS of BEAS-2B and MH-S cells.(F and G) Western blot analysis and quantitative results of GPX4, SLC7A11, and HO-1 of BEAS-2B cells.(H and I) Western blot analysis and quantitative results of GPX4, SLC7A11, and HO-1 of MH-S cells.(J and K) Western blot ana

PM2.5 triggered ferroptosis in the lungs of mice(A) Volcano plot of RNA-seq results of lung tissues exposed to filtered air or PM2.5 showing statistics of upregulated and downregulated genes (DEGs; log2 FC ≥ 1 or ≤ −1, respectively; *p* ≤ 0.05) (*n* = 5 mice per group).(B) Principal-component analysis (PCA) of RNA-seq data for Ctrl and PM2.5 mice.(C) A scatterplot of the top 15 KEGG pathways enriched among the DEGs.(D and E) MDA and SOD levels in lung tissues of mice following three different exposure routes (*n* = 6 mice per group).(F) qPCR analysis of GPX4, SLC7A11, Nrf2, and FTH1 in lung ti

Full text

Introduction

Air pollution continues to be the most pronounced environmental health challenge globally. Poor air quality is accountable for over 6.7 million premature deaths each year, most of which are due to respiratory and cardiovascular causes.1 Particulate matter (PM), a complex mixture of gaseous, liquid, and solid substances, is identified as a primary cause of air pollution. Based on the latest evidence, 94% of the global population resides in areas with PM2.5 exposure exceeding the WHO air quality guidelines (less than 5 μg/m^3^ annually), which increases mortality rates by 4%. Among these, about 36% of the world population is directly exposed to concentrations above 35 μg/m^3^, leading to a 24% increase in mortality rates.2 The 2021 global burden of disease report indicated that PM pollution remains the leading contributor to the global disease burden, contributing 231.5 million of disability-adjusted life-years (DALYs), despite improvements in the risk-attributable burden associated with household air pollution during the 2000–2021 period.3 Furthermore, the health impacts attributed to pollution disproportionately affect heavily populated countries. In China, over 99% of the population faces unsafe PM2.5 levels (over 5 μg/m^3^), and 53% faces hazardous levels (over 35 μg/m^3^).2 Accordingly, the impact of air pollution on health remains a significant concern that should not be overlooked, despite the implementation of effective governance in recent years.

Mountains of studies have established the specific respiratory health effects associated with exposure to PM2.5.4^,^5^,^6 Consequently, a growing body of research in recent decades has focused on the mechanisms by which PM2.5 induces lung injury utilizing animal models. However, there is a notable lack of consensus on the exposure procedures employed in these studies, presenting substantial challenges to the reproducibility of the experiments. Furthermore, much less is understood regarding the difference among these methods and the potential factors influencing their toxicity, which needs to be elucidated. In addition, most research has predominantly focused on the health effects of the lungs, given that the lungs are widely regarded as the primary target organs.7 The defensive role of the upper respiratory tract against foreign particles has been largely neglected. In fact, the upper respiratory tract is integral to the defense and clearance of exogenous invaders. Therefore, a comprehensive investigation into the effects of PM2.5 on both the upper and lower respiratory tracts is warranted.

The adverse health effects of PM are primarily determined by particle size and constituents. PM is classified by size into ultrafine particles (less than 0.1 μm [PM0.1]), fine particles (ranging from 0.1 to 2.5 μm [PM2.5]), and coarse particles (greater than 2.5 μm up to 10 μm [PM10]). PM can contain various aerosol constituents, including sulfates, ozone, nitrates, and volatile organic compounds, along with heavy metals and polycyclic aromatic hydrocarbons (PAHs).8 These components are critical contributors to PM2.5-induced oxidative stress, DNA damage, and cell death.9^,^10^,^11 The involvement of PM, especially PM2.5, in the onset and exacerbation of various diseases has been extensively documented,12^,^13^,^14^,^15 since they can escape from the mucous barrier and enter the alveolar ducts and capillaries, even disseminate throughout the human body via the circulatory system. The lung defense system, which is responsible for blocking and removing inhaled debris, can be compromised by PM2.5. Understanding the mechanisms of damage leading to diseases such as lung cancer or pneumonia from PM2.5 exposure is essential.

Ferroptosis is an oxidative stress-induced, iron-dependent form of programmed cell death. Iron accumulation, elevated lipid peroxidation, and inability to effectively reduce lipid peroxides are the three major pillars of engaged in the ferroptotic cascade.16 It has been found that the extensive release of oxidized lipid mediators following ferroptosis can trigger inflammatory responses and airway remodeling in the lungs, thereby contributing to the pathogenesis of various pulmonary disorders.17^,^18 Pharmacological modulation of ferroptosis has emerged as a promising therapeutic strategy for the treatment of lung cancer, COPD, and pulmonary fibrosis in preclinical animal models.19^,^20 Recent studies have also revealed that diesel exhaust PM2.5 induced mitochondrial dysfunction and excessive production of mitoROS, leading to ferroptosis and ferritinophagy in pulmonary cells, and blocking ferroptosis has been shown to mitigate PM2.5-induced airway inflammation and oxidative stress.21^,^22 Moreover, the oxidative potential of PM2.5 is regarded to increase the sensitivity of pulmonary epithelial cells to oxidative stress and ferroptosis, which can be reduced by the use of radical scavengers.23 Nevertheless, the specific role of ferroptosis in PM2.5-induced lung risks and the identification of key targets remain unclear.

This study aimed to compare the respiratory toxicity of PM2.5 via three different exposure routes in mice: intranasal, intratracheal, and whole-body, then to investigate the underlying mechanisms by which PM2.5 enters the lungs and causes dysfunction, utilizing two types of cellular model (airway epithelial cells and alveolar macrophages), with a focus on epithelial barrier integrity and macrophage phagocytic ability. The findings revealed a correlation between the biological uptake of PM2.5 and respiratory toxicity, wherein increased pulmonary uptake and retention of PM2.5 were associated with more severe lung injuries. This suggests that the number of barriers PM2.5 encounters during transit significantly influences its uptake and retention. Moreover, the lung defense system, including airway epithelial barriers and phagocytic ability of alveolar macrophages, was compromised by PM2.5 induction, in which ferroptosis-mediated oxidative stress played a crucial role. This research offers valuable insights into the molecular pathways involved in PM2.5-induced lung damage. The identified downregulation of anti-ferroptotic genes, such as FTH1, provides potential targets for therapeutic intervention to mitigate the adverse effects of PM2.5 on lung health.

Results

PM2.5 induced an acute inflammation in the lung

All mice were exposed to PM2.5 via three different pathways once every other day, and lung function was assessed within 24 h following PM2.5 exposure and after the cessation of exposure, with this procedure being repeated three times (Figure 1A). The gross anatomy of the lungs exhibited distinct black PM2.5 particles stuck in the bronchus and lung tissues of mice following various exposure patterns, with a notable prevalence in the TE group (Figure 1B). Pathological examination showed extensive acute and abundant inflammatory cell infiltration in the bronchioles and alveoli following PM2.5 exposure via both NE and TE methods, with a particularly pronounced effect in the TE group. Of note, distinct black PM2.5 aggregates were observed to accumulate in the alveolar spaces surrounding the bronchus. However, negligible inflammatory cells and PM2.5 particles were observed under the whole-body exposure (WE) (Figures 1C and 1D). Subsequently, the inflammatory cytokines were assessed, revealing significant increases in IL-6 levels in both serum and BALF following PM2.5 exposure via TE. NE also results in a substantial elevation of both TNF-α and IL-6 in serum and BALF. In contrast, exposure through WE did not lead to elevated levels of either IL-6 or TNF-α, aligning with the observed lung pathological manifestations (Figures 1E and 1F).Figure 1. Acute inflammation in the lungs of mice following three different exposure protocols(A) Schematic overview of the experimental design. Mice were exposed to PM2.5 exposure via three different methods: single intranasal instillation (NE), intratracheal instillation (TE), or whole-body exposure (WE) (blue arrows), followed by a one-day cessation, and this cycle was repeated three times. After the third exposure, all mice were euthanized. Non-invasive pulmonary function tests were conducted after each exposure and cessation period (black arrows).(B) Gross observation of lung tissues.(C) Histological images of lung sections stained with H&E following the final exposure. Scale bars: upper, 500 μm; below, 50 μm. The infiltration of inflammatory cells is denoted by black arrows.(D) Inflammation score of lung tissues.(E and F) IL-6 and TNF-α levels in BALF and serum are measured by ELISA within 24 h after the final exposure. ^∗^p < 0.05 and ^∗∗^p < 0.01.(G–J) Lung volume parameters, including frequency (f), minute volume (MV), tidal volume (TV), and expiratory time (Te).Data are presented as mean ± SEM (n = 6 mice per group). ^∗^p < 0.05 and ^∗∗^p < 0.01 (TE vs. Ctrl); ^#^p < 0.05 and ^##^p < 0.01 (NE vs. Ctrl), as determined by two-way ANOVA followed by Dunnett’s multiple comparison test.

For lung function, within 24 h following exposure to PM2.5 through TE, there was a significant decrease in f, MV, and TV, alongside an increase in Te, indicating a marked reduction in lung volume. However, this reduction reverted to baseline levels after a one-day cessation of PM2.5 exposure. This pattern of “decreased lung volume upon exposure and recovered upon cessation” was observed three times, though it diminished with prolonged exposure and was not prominent until the third exposure cycle. In contrast, during the initial two instances of NE exposure, no significant reduction in lung volume was observed until later stages. WE exposure to PM2.5 did not result in any changes in these indices at any exposure time (Figure 1G). Other lung function indices, including airway flow, as represented by EF50 and Penh, and lung elasticity, as indicated by peak inspiratory flow (PIF) and peak expiratory flow (PEF), exhibited a similar trend (Figure S1). These results suggest that acute exposure to high concentration of PM2.5 leads to significant lung function impairment and inflammatory response, particularly through the TE pathway.

PM2.5 induces mucus hypersecretion and epithelial barrier impairment in the upper and lower respiratory tracts

The upper respiratory tract, primarily comprising the nasal cavity and trachea, serves as the initial target of contact for foreign particles and plays a crucial role in the defense against and clearance of PM2.5. Therefore, histopathological examinations of the nasal cavity and trachea were conducted following various exposure scenarios. Consequently, there was a marked inflammatory cell influx in the lamina propria of the nasal mucosa post-exposure to PM2.5. Of note, TE and WE resulted in a greater increase in inflammatory cells compared to NE (Figure 2A). Furthermore, AB/periodic acid-Schiff (PAS) staining showed a significant increase in mucosubstances across all treatments, suggesting an enhanced physical defense in the upper respiratory tract to eliminate foreign particles. Interestingly, WE induced the highest increase in mucosubstances within the mucous glands of the trachea compared with other treatments, indicated by the intensity of the darkened blue color (Figure 2B).Figure 2. Changes of the epithelial barrier in the upper and lower respiratory tracts of mice following various exposure patterns(A) Representative micrographs of nasal and tracheal tissues stained with H&E. Scale bars: 50 μm. Goblet cell metaplasia is denoted by black arrowheads, ciliary structure by green arrowheads, and inflammatory cell influx by red arrowheads.(B) Representative micrographs of nasal and tracheal tissues stained with AB-PAS. Scale bars: upper, 50 μm; below, 100 μm. Nasal mucosubstances are denoted by black arrows, and mucous gland hyperplasia is indicated by orange arrows.(C–F) Representative immunofluorescence images depicting tight junction markers (ZO-1, occludin, and E-cadherin) and quantitative analysis of relative fluorescence intensities in the lungs of mice following the final exposure. Scale bars: 50 μm.(G–J) Representative immunofluorescence images and quantitative analysis of mucins and ciliary proteins (MUC5AC, MUC5B, CLCA1). Scale bars: 50 μm.Data are presented as mean ± SEM (n = 6 mice per group). ^∗∗^p < 0.01 as determined by two-way ANOVA followed by Dunnett’s multiple comparison test.

In addition, the epithelial barrier plays a crucial role in protecting the lungs against the penetration of noxious particles and pathogens. As previously noted, exposure to inhaled air pollutants directly or indirectly causes airway epithelial barrier dysfunction.24 In this study, a significant reduction in the expression of tight junction proteins, such as ZO-1 and occludin, as well as adherens junction proteins, including E-cadherin, was observed in the bronchoalveolar epithelium of mice following exposure to PM2.5, irrespective of the exposure patterns (Figures 2C and 2D). Further, the disruption of barrier function is invariably associated with mucociliary dysfunction, characterized by mucus hypersecretion, ciliary loss, and impaired ciliary motility, all of which contribute to airway obstruction.25 In comparison to the control group, there was a significant upregulation of MUC5AC and MUC5B expression in the airway epithelial cells of mice across all exposure patterns (Figures 2E and 2F). These findings suggest that the barrier functions of epithelial cells were compromised following PM2.5 exposure.

PM2.5 impaired the epithelial barrier partially by promoting ferroptosis in BEAS-2B cells

To elucidate the adverse effect of PM2.5 on the airway epithelial barrier, we developed an in vitro model utilizing human airway epithelial cells (BEAS-2B) subjected to varying concentrations of PM2.5 for durations ranging from 30 min to 12 h. The western blot data showed a notable reduction in the expression of ZO-1, occludin, and E-cadherin in the human airway epithelial cells (BEAS-2B) up to 3 h, persisting for at least 6 h in a dose-dependent manner (Figures 3A and 3B), indicating that PM2.5 induced a significant impairment in the functionality of the epithelial barrier system.Figure 3PM2.5 impaired the epithelial barrier partially by promoting ferroptosis in BEAS-2B cells(A and B) Western blot analysis and quantitative results of ZO-1, occludin, and E-cad in BEAS-2B cells exposed to varying concentrations of PM2.5 (10, 20, 50, and 100 μg/cm^2^) over different time periods (1 h, 3 h, and 6 h).(C) Volcano plots of differentially expressed genes (DEGs)in BEAS-2B cells from PM2.5 exposure and Control groups in a dataset ([GSE155616](GSE155616)). Red points represented upregulated genes and blue points represented downregulated genes filtered based on fold change (|logFC|) ≥1 and p < 0.05.(D) KEGG enrichment bubble plots of DEGs in two comparisons. The color of bubbles indicated the value of p value, and the size of bubbles stated the number of counts.(E) GO enrichment barplot of these DEGs.(F–H) MDA, SOD, and GSH levels of BEAS-2B cells treated by PM2.5 (20 and 50 μg/cm^2^) with or without Fer-1 (5 μM) for 6 h.(I and J) Western blot analysis and quantitative results of GPX4, SLC7A11, and HO-1.(K and L) Western blot analysis and quantitative results of ZO-1, occludin, and E-cadherin.Data are presented as mean ± SEM (n = 4 independent experiments). ^∗^p < 0.05 and ^∗∗^p < 0.01 as determined by two-way ANOVA followed by Dunnett’s multiple comparison test.

To further understand the mechanisms of PM2.5 induced epithelial barrier impairment in BEAS-2B, we searched the GEO database to obtain relevant mRNA sequencing data. As shown in Figure 3C, 221 significant differential genes were exhibited between the control and PM2.5 groups, of which 106 were upregulated and 115 were downregulated. Through KEGG enrichment analysis, we found that PM2.5 may play its role in promoting lung injury through cytokine-cytokine receptor interaction, lipid and atherosclerosis, PI3K-Akt signaling pathway, TNF signaling pathway, chemical carcinogenesis-reactive oxygen species, and ferroptosis (Figure 3D). Subsequent GO analysis showed that PM2.5 primarily enriched categories associated with positive regulation of ferroptosis, cell adhesion, extracellular matrix, ferritin complex, and iron ion binding (Figure 3E). These observations indicate that PM2.5 appears to destroy the airway epithelial barrier by promoting ferroptosis.

Subsequently, we investigated the effect of Fer-1, a ferroptosis inhibitor, to verify the role of ferroptosis in PM2.5-induced disruption of the epithelial barrier. The results demonstrated that along with the increase of MDA, the levels of GSH and SOD were significantly decreased following PM2.5 exposure in BEAS-2B. However, Fer-1 reversed the effects of PM2.5 on MDA, GSH and SOD (Figures 3F–3H). Consistently, Western blot and qPCR analysis indicated that Fer-1 addition led to an up-regulation of GPX4, and SLC7A11 expression (Figures 3I, 3J, and S2). Furthermore, we confirmed that the PM2.5-induced decreases of epithelial barrier markers, including ZO-1, occludin, and E-cadherin, were also elevated by Fer-1 in BEAS-2B (Figures 3I–3K). Taken together, these results suggest that ferroptosis is the predominant pathway contributing to PM2.5-induced injury of the airway epithelial barrier.

PM2.5 triggered ferroptosis to promote inflammation in the alveolar macrophages

In addition to the epithelial barrier systems, macrophages are also normally recognized as the first-line defenders to clear the invading particles.26 Figure 4A illustrates a marked increase in CD68^+^ macrophages, with the TE group exhibiting the highest increase. Moreover, a significant accumulation of PM2.5 particles was observed surrounding the macrophages, consistent with the observations in the BALF (Figures S3A–S3C), suggesting an activation of the alveolar macrophages’ defensive response to eliminate foreign particles. In the MH-S cells, a typical model of alveolar macrophages, substantial aggregates of PM2.5 particles could be observed within the cytoplasm, with some adhering to the plasma membrane, indicating the high phagocytosis of PM2.5 by MH-S cells. However, with a prolonged PM2.5 exposure for 6 h, there was a noticeable reduction in both filopodia protrusions and fluorescence particles in the cells, suggesting a disruption of phagocytic function induced by PM2.5 (Figures 4B–4D). Meanwhile, a significantly dose-dependent decrease in cell viability and an increase in the secretion of inflammatory factors (IL-6 and TNF-α) were observed (Figures S3D–S3F). These findings suggest that acute PM2.5 exposure can activate the defensive capability of alveolar macrophages, but concurrently diminishes their phagocytic function and promotes inflammation.Figure 4PM2.5 triggered ferroptosis and promoted inflammation in MH-S cells(A) Immunofluorescence analysis and quantification of CD68 expression (white arrows) in the lung tissues of mice subjected to various exposure patterns. Scale bars: 50 μm. Red arrows highlight the particle aggregates.(B) Confocal laser scanning microscopy images illustrating the phagocytic capacity of MH-S following PM2.5 exposure at various time intervals. Yellow arrowheads denote particle aggregates, and white arrowheads indicate filopodia protrusions. Scale bars: 50 μm.(C) The average number of pHrodo green Zymosan bioparticles ingested per cell.(D) The percentage of cells containing pHrodo green Zymosan bioparticles.(E–G) Volcano plots, KEGG enrichment, and GO analysis of DEGs in macrophages from PM2.5 exposure and Control groups in a dataset ([GSE294723](GSE294723)).(H) Level of MDA in MH-S treated with PM2.5 (20 and 50 μg/cm^2^) with or without the addition of Fer-1 (5 μM) for 6 h.(I) qPCR analysis of GPX4, SLC7A11, Nrf2, and SOD2.(J and K) IL-6 and TNF-α levels in MH-S detected by ELISA.(L–N) Immunofluorescence images and quantitative assessments of the phagocytic capacity of MH-S treated with PM2.5 (50 μg/cm^2^) with or without Fer-1 (5 μM) for 6 h. Scale bars: 50 μm.Data are presented as mean ± SEM (n = 4 independent experiments). ^∗^p < 0.05 and ^∗∗^p < 0.01 as determined by two-way ANOVA followed by Dunnett’s multiple comparison test.

To further substantiate the potential adverse effects and underlying mechanisms of PM2.5 on macrophages, we also searched the GEO database related to the PM2.5-exposed macrophages, and obtained 39 differentially expressed genes (DEGs) (Figure 4E). Interestingly, these 39 genes were notably enriched in the ferroptosis and mineral absorption (Figure 4F), similar to the previous data observed in BEAS-2B. In the GO enrichment analysis, PM2.5 specifically affected the homeostasis of copper, zinc, and iron ions, as well as the detection of oxygen, highlighting the involvement of ferroptosis and oxidative stress in the regulation of macrophages’ function upon PM2.5 exposure (Figure 4G). Furthermore, treatment with Fer-1 markedly mitigated PM2.5-induced MDA accumulation (Figure 4H) and restored the expressions of the GPX4, SLC7A11, Nrf2, and SOD2 ferroptosis genes (Figure 4I). We further checked key mediators involved in inflammation, and observed both IL-6 and TNF-α were upregulated in PM2.5-exposed MH-S, but reduced after Fer-1 treatment (Figures 4J and 4K). Additionally, the macrophage phagocytosis assay exhibited higher fluorescence particle uptake in the MH-S after Fer-1 treatment (Figures 4L–4N). These findings collectively suggest that PM2.5 disrupts macrophage phagocytosis and elicits a substantial inflammatory response, which can be reversed through the inhibition of ferroptosis.

FTH1 was a key determinant for PM2.5-induced ferroptosis in BEAS-2B and MH-S

Furthermore, a PPI network was developed for the 221 identified targets from the BEAS-2B dataset, of which 28 central targets were identified. Among these targets, ferritin heavy chain 1 (FTH1) has the highest degree value (Figure 5A). As expected, western blot revealed that PM2.5 resulted in a significant reduction in the expression of FTH1 in a dose-dependent manner, but Fer-1 treatment reversed its expression in both BEAS-2B and MH-S (Figures 5B and 5C).Figure 5PM2.5 resulted in ferroptosis by upregulating FTH1 in BEAS-2B and MH-S(A) Hub DEGs were identified from the PPI network using Cytoscape, based on the [GSE155616](GSE155616) dataset.(B and C) Western blot analysis of FTH1 in PM2.5-induced BEAS-2B and MH-S with or without Fer-1 addition.(C–E) Flow cytometry results of intracellular ROS of BEAS-2B and MH-S cells.(F and G) Western blot analysis and quantitative results of GPX4, SLC7A11, and HO-1 of BEAS-2B cells.(H and I) Western blot analysis and quantitative results of GPX4, SLC7A11, and HO-1 of MH-S cells.(J and K) Western blot analysis of ZO-1, occludin, and E-cadherin in BEAS-2B.Data are presented as mean ± SEM (n = 3 independent experiments). ^∗^p < 0.05 and ^∗∗^p < 0.01 as determined by two-way ANOVA followed by Dunnett’s multiple comparison test.

Next, BEAS-2B and MH-S cells were transfected with the FTH1 plasmid with or without PM2.5. The data showed that the intracellular reactive oxygen species (ROS) level of the cells transfected by the FTH1-overexpressed plasmid was remarkably lower than that in untransfected cells (Figures 5D and 5E). However, higher GPX4, SLC7A11, and HO-1 were observed after FTH1 overexpression (Figures 5F–5I and S4). Moreover, overexpression of FTH1 significantly abrogated PM2.5-induced epithelial barrier impairment in BEAS-2B (Figures 5J and 5K) and inflammatory response in MH-S (Figure S4). Taken together, these results suggest that FTH1 was a critical regulator for PM2.5-induced ferroptosis in BEAS-2B and MH-S.

PM2.5 triggered ferroptosis in the lungs of mice

For the transcriptional differences between filtered air and PM2.5-exposed lungs in mice, 3326 DEGs were identified by RNA-sequencing (RNA-seq), of which 2059 were increased and 1267 were decreased (Figures 6A and 6B). KEGG analysis revealed the DEGs enrichment mainly in pathways related to oxidative stress, such as chemical carcinogenesis-reactive oxygen species, oxidative phosphorylation, glutathione metabolism, and ferroptosis, followed by adhesion-related signaling pathways, such as ECM-receptor interaction and focal adhesion pathways (Figure 6C), which further suggested a critical role of epithelial barriers and ferroptosis in regulating PM2.5-induced lung injury.Figure 6PM2.5 triggered ferroptosis in the lungs of mice(A) Volcano plot of RNA-seq results of lung tissues exposed to filtered air or PM2.5 showing statistics of upregulated and downregulated genes (DEGs; log2 FC ≥ 1 or ≤ −1, respectively; p ≤ 0.05) (n = 5 mice per group).(B) Principal-component analysis (PCA) of RNA-seq data for Ctrl and PM2.5 mice.(C) A scatterplot of the top 15 KEGG pathways enriched among the DEGs.(D and E) MDA and SOD levels in lung tissues of mice following three different exposure routes (n = 6 mice per group).(F) qPCR analysis of GPX4, SLC7A11, Nrf2, and FTH1 in lung tissues of mice following three different exposure routes (n = 6 mice per group).(G) Histological images and inflammation score results of lung sections stained with H&E following TE-PM2.5 exposure with or without Fer-1 (2 mg/kg) pretreatment (n = 6 mice per group). Scale bars: 50 μm.(H–K) Western blot analysis and quantitative results of occludin, E-cadherin, and GPX4 in lung tissues (n = 3 mice per group).Data are presented as mean ± SEM. ^∗^p < 0.05 and ^∗∗^p < 0.01 as determined by two-way ANOVA followed by Dunnett’s multiple comparison test.

In vivo, a significant increase in MDA and decline in SOD were detected in the lungs of mice exposed to PM2.5 through the three forementioned exposure patterns (Figures 6D and 6E). The expressions of GPX4, SLC7A11, HO-1, and FTH1 were notably down-regulated after PM2.5 induction, with the TE routes exhibiting the most prominent effect (Figure 6F). Then, based on the TE-PM2.5 mouse model, we further evaluated the impact of Fer-1 to verify the role of ferroptosis in PM2.5-induced lung injury. The results showed that Fer-1 treatment mitigated the pathological damage of lung tissues caused by PM2.5 exposure (Figure 6G) and exhibited noticeable anti-oxidative and anti-inflammatory properties (Figure S5). To confirm the protective effect of Fer-1 on the epithelial barrier, we assessed ZO-1 and occludin by western blot and found marked decreasing trends of ZO-1 and occludin induced by PM2.5, which were significantly abolished by Fer-1 treatment (Figures 6H–6K). Collectively, our results confirmed the role of ferroptosis in mediating PM2.5-induced pulmonary injury, encompassing oxidative stress, inflammatory response, and epithelial barrier disruption.

Discussion

This study sought to evaluate the potential for PM2.5 particles to penetrate the lungs and cause acute pulmonary toxicity through three respiratory exposure routes: intratracheal, intranasal, and whole-body. Given the focus on the effects of acute PM2.5 exposure, the dose of PM2.5 was determined based on the daily highest concentration of PM2.5 (350 μg/m^3^) in winter from 2018 to 2023 in China. This concentration was then converted to a total daily dosage by multiplying the total respiratory volume of adult mice (0.13 m^3^), resulting in a daily dose of 4.55 mg, which has been previously reported to assess the effects of acute PM2.5 exposure.27 Moreover, we utilized young adult (10–12 week-old) male mice to minimize confounding hormonal variability and to establish a clear baseline model, free from age-related complications, for assessing PM2.5 toxicity. In this study, acute PM2.5 exposure was found to cause histological damage to both the upper and lower respiratory systems, as indicated by mucus hypersecretion, inflammatory cell infiltration, and ciliary disruption. Additionally, we observed that PM2.5 impaired the junction barrier in airway epithelial cells (BEAS-2B) and diminished the phagocytic capacity of alveolar macrophages (MH-S). Mechanistically, FTH1-mediated ferroptosis was demonstrated to participate in the PM2.5-induced epithelial barrier disruption and inflammation, which underscores a promising therapeutic approach in PM2.5-induced lung injury therapy.

As a main indicator for evaluating respiratory diseases, lung function change following PM exposure has been widely recognized, but the findings remain contradictory. Several longitudinal cohort studies reported that short- and long-term PM2.5 exposure was negatively associated with forced expiratory volume in 1 s (FEV_1_) and forced vital capacity (FVC) in adults, in particular long-term exposure.28^,^29^,^30^,^31 A large cohort study containing 3,262 adult participants indicated that even a single day of PM2.5 exposure was linked to reduced lung function.32 However, these results were disputed by an eighteen-year follow-up study which suggested that unlike O3, PM2.5 exposure over follow-up was not significantly correlated with emphysema progression and lung function decline.33 Meier Reto also found no significant changes in lung function within 15 h after traffic-related PM2.5 exposure.34 In this study, we observed that a single high-level exposure to PM2.5 resulted in considerable airway obstruction and ventilation defects, while reverting to normal levels after every one-day cessation. We proposed that the acute effects of PM2.5 on the lung primarily involve inflammation and oxidative stress, which may not be enough to affect lung function. However, whether lung function is valuable for evaluating the acute lung toxicity of PM2.5 remains to be seen.

In this study, we observed that both intratracheal and intranasal instillation of PM2.5 increased the levels of inflammatory cytokines in serum and BALF and induced lung histopathological injuries. The severity of lung injury caused by intratracheal instillation was significantly greater than that caused by intranasal instillation. Moreover, obvious deposition of PM2.5 in the lungs was observed via both exposure routes, with deposition being at least twice as high after intratracheal than after intranasal exposure. Unexpectedly, PM2.5 administered via the whole-body route also reached the lungs, however, the quantity was minimal and insufficient to induce lung toxicity. Nevertheless, WE mainly induced histological damage in the nasal cavity and trachea, similar to the other two routes. These findings suggest that exposure via any route can potentially harm the respiratory system, with lung toxicity being most evident following intratracheal exposure due to the greatest amount of toxic substance reaching the lungs. Similar research has indicated that the intratracheal route is more effective in delivering PM to the lungs than the intranasal route.35 This enhanced efficacy may be attributed to the direct administration of particles in the proximity of lungs escaping several barriers, such as the skin, intranasal, and mucociliary barriers.

The epithelial tissue acts as a barrier to protect the body from environmental stress, infections, and allergens, mainly including a robust physical barrier in the skin, continuous particle clearance to maintain efficient gas exchange in the lungs, and extensive nutrient and water exchange in the gastrointestinal system.24 Notably, the respiratory barrier is the first line to encounter inhaled pollution, contributed by mucociliary escalators, intercellular protein junctions, and secreted antimicrobial products.25 The link between environmental exposures and the epithelial barrier is now becoming increasingly clear. Previous studies have documented that airway epithelial cells exposed to PM exhibited compromised barrier integrity and airway dysfunction with decreased tight junction (TJ) proteins, elevated fluorescein tracer permeability, and reduced transepithelial electrical resistance.36^,^37 In a murine model, PM2.5 exposure has been shown to suppress E-cadherin expression in the lungs.38 Our results agree with these studies in finding that both mice and airway epithelial cells subjected to PM2.5 had significant epithelial barrier disruption, as assessed by downregulated TJ proteins, and increased mucus secretion. Likewise, PM2.5 exposure damaged the upper respiratory tract proven by disrupted cilia, mucus hypersecretion, and inflammatory cell infiltration in the nasal cavity and trachea, even via the whole-body route. These findings specifically indicated the adverse impacts of PM2.5 on the respiratory epithelial barriers. However, there is ongoing debate regarding the limitations of in vitro models using 2D monocultures of cell lines, as they fail to replicate the complex microenvironment of the lung. Consequently, recent studies utilizing the lung-on-a-chip models have further elucidated the correlation between PM2.5 exposure and the disruption of the alveolar-capillary barrier.39^,^40

In addition to the airway epithelial barrier, macrophages are normally recognized as the first line of defense against invading particles by playing an important role in particle transport and clearance.41 Upon inhalation into the respiratory tract, PM2.5 can enter the lung deeply and then be endocytosed by macrophages. A recent study demonstrated that massive particles were visualized inside macrophages upon incubation for 24 h and incurred severe phospholipid membrane disruption in macrophages, impairing the innate immune function.42 Moreover, the intratracheally instilled particle-laden alveolar macrophages can translocate both ultrafine and fine particles from the lungs to extrapulmonary organs.43 Our previous data revealed that macrophages would undergo autophagy-lysosomal flux disruption and increased inflammation after endocytosis of PM2.5.44 However, rather limited knowledge has been obtained to understand the effect of PM2.5 on the phagocytic function of macrophages. It has been indicated by a bronchoscopy that short-term exposure to wood smoke (409 μg/m^3^) reduced cell viability and impaired alveolar macrophage phagocytic capacity in healthy humans.45 This study observed a significant accumulation of PM2.5 in the BALF and lungs, mainly localized within macrophages. Moreover, after 6 h of incubation, the alveolar macrophages (MH-S) showed markedly impaired phagocytic capacity, as evidenced by compromised cytoskeleton, diminished filopodia protrusions, and minimal uptake of fluorescence particles by MH-S. The mechanisms of cytotoxicity to macrophages caused by PM may be ascribed to oxidative stress, cell membrane barrier damage, or inflammatory response.46 Our prior research has indicated the significantly promoted effect of PM2.5 on inflammation,44 yet the impact of intruded PM2.5 on the phagocytic ability of macrophages remains to be fully elucidated.

Surface-stabilized environmentally persistent free radicals on PM play a critical role in the production of ROS. After being taken into pulmonary cells, PM2.5 activates oxidases and metabolic enzymes and causes mitochondrial dysfunction, leading to ROS overproduction and disturbing the intracellular antioxidant defense system.47 Oxidative stress induced by PM2.5 is attributed to its adherent constituents, notably metals and PAHs. It is well-established that reductive metals, such as iron (Fe) and copper (Cu), generate reactive radicals such as superoxide radicals and nitric oxide free radicals in biological systems via Fenton reactions.48 PAHs and their derivatives react with molecular oxygen and other mediators to produce ROS.49 The PM2.5 samples collected in this study were found to contain abundant metal elements and PAHs (Table S1), which contributed to the excessive production of ROS, thereby reducing the activity of antioxidant enzymes in the lung. Furthermore, excessive free radicals might further attack and oxidize phospholipids in the membrane to trigger ferroptosis, which is characterized by the accumulation of lipid peroxides and iron-dependent ROS, leading to oxidative damage and, ultimately, cell death.16 In the present study, we examined the activation of ferroptosis-related signaling markers in airway epithelial cells, alveolar macrophages, and lung tissues from the mouse model exposed to PM2.5 through transcriptome sequencing. Notably, intratracheal exposure to PM2.5 resulted in the most pronounced ferroptosis along with lipid peroxidation and GPX4 inhibition, consistent with the pathological injury. Moreover, ferroptosis inhibitor Fer-1 treatment directly reversed not only the pathological injury in the lungs but also mitigated oxidative stress, epithelial barrier impairment, and inflammation induced by PM2.5. This is consistent with emerging evidence that chronic, low-dose PM2.5 exposure can also induce ferroptosis, suggesting it as a pivotal pathway in the toxicity of PM2.5.50^,^51 However, further investigation is needed to elucidate the potential roles of other cell death mechanisms and confirm the predominance of ferroptosis. Iron transport and overload are fundamental mechanisms that exogenously drive ferroptosis. Transferrin (TF) and its receptor work together to facilitate ferroptosis by mediating iron import into the cell. Conversely, ferritin, composed of two types of polypeptide chains, ferritin heavy chain (FTH1) and ferritin light chain (FTL), plays an important role in iron storage and the maintenance of iron homeostasis, thereby protecting against oxidative stress and iron toxicity. Notably, FTH1 catalyzes the conversion of iron from its ferrous form (Fe^2+^) to the ferric form (Fe^3+^), enabling iron to be stored in an inert form within the ferritin shell.16^,^52 It has been demonstrated that inhibiting FTH1 reduces ferritin’s ability to store iron in a stable form, leading to labile iron pool overload. The excess iron can then catalyze the Fenton reaction within mitochondria, generating excessive ROS and directly promoting lipid peroxidation, ultimately triggering ferroptosis and oxidative stress.53 Glutathione peroxidase 4 (GPX4) is recognized as the central initiator and mediator of ferroptosis by catalyzing the reduction of phospholipid hydroperoxides and protecting cell membranes from lipid peroxidation. Exposure to fine dust has been documented to decrease GPX4 expression, thereby enhancing lipid peroxidation and inducing oxidative stress, which subsequently results in the activation of Nrf2 and upregulation of HO-1.21^,^22^,^54 In addition, the expression of solute carrier family 7A member 11 (SLC7A11), another reliable marker of ferroptosis, has been shown to decrease in PM2.5-induced cardiomyocytes, an effect that is reversed by Fer-1 pretreatment.55 In this study, our network analysis identified FTH1 as a key node in regulating PM2.5-induced lung injury. The protein expression of FTH1 exhibited a marked decreasing trend following PM2.5 exposure in both the mice lungs and cells. And there were significant changes in other ferroptosis-related indicators, such as the increase of oxidative damage markers ROS, MDA, HO-1, and TF, alongside the decrease of antioxidants SOD, GSH, GPX4, and SLC7A11. Notably, FTH1 overexpression significantly restricted PM2.5-induced ferroptosis, airway epithelial barrier impairment, and inflammatory responses. Collectively, these findings suggest that PM2.5 induces ferroptosis by downregulating the expression of the iron storage protein FTH1 in the lungs.

In summary, our study provides a foundational animal model for examining the health risks associated with air pollution and identifies two pivotal factors (the epithelial barrier and macrophage function) in particle transport and clearance, although further research is necessary. Moreover, our findings demonstrated that the environmental pollutant PM2.5 induces lung injury, at least in part, by triggering FTH1-dependent ferroptosis, which could offer an earlier therapeutic approach for treating pulmonary diseases, particularly under conditions of environmental pollution.

Limitations of the study

Despite these findings, several limitations of this study should be acknowledged. First, our study was inspired by comparing the lung toxicity of PM2.5 through different exposure routes. Integrating comparative transcriptomic profiling across these models could provide a more mechanistic and systematic insight into how varying PM2.5 exposure paradigms distinctly modulate pathogenic pathways. Second, the study lacks single-cell RNA sequencing data to precisely identify the specific cellular subtypes involved in PM2.5-induced lung injury. Last, this study exclusively used male mice to avoid potential confounding effects of the estrous cycle on the experimental outcomes. As a result, whether these findings can be generalized to female subjects remains unclear.

Resource availability

Lead contact

Requests for further information and resources should be directed to and will be fulfilled by the lead contact, Jing Wang (wjing2431@126.com).

Materials availability

This study did not generate new unique reagents.

Data and code availability

• •This paper analyzes publicly available data [GSE155616](GSE155616) and [GSE294723](GSE294723).
• •This paper does not report the original code.
• •Any additional information is available from the lead contact upon reasonable request.

Acknowledgments

This work was financially supported by the National Natural Science Fund of China (grant nos 82574738, 82274495, and 82374263), Key Scientific Research Projects of Higher Education Institutions in Henan Province (26A360002).

Author contributions

Conceptualization, J.W. and J.L.; methodology, H.Y., J.W., Y.L., H.D., and Y.D.; data curation, H.Y., J.W., H.D., and T.L.; formal analysis, H.Y. and Y.D.; investigation, T.L.; writing—original draft, H.Y. and J.W.; writing—review & editing, J.W.; funding acquisition, Y.L., T.L., J.W., and J.L.; supervision, T.L.

Declaration of interests

The authors declare no competing interests.

STAR★Methods

Key resources table

REAGENT or RESOURCESOURCEIDENTIFIERAntibodiesAnti-ZO-1ProteintechCat# 21773-1-AP; RRID: AB_10733242Anti-E-cadherinProteintechCat# 20874-1-AP; RRID: AB_10697811Anti-OccludinProteintechCat# 27260-1-AP; RRID: AB_2880820Anti-GAPDHProteintechCat# 60004-1-Ig; RRID: AB_2263076Anti-CLCA1AbcamCat# ab180851; RRID: AB_2722611Anti-MUC5BCloud-CloneCat# PAA684Mu01Anti-MUC5ACCloud-CloneCat# PAA756Mu01Anti-CD68santa cruzCat# sc-20060; RRID: AB_627158Anti-GPX4AbcamCat# ab125066; RRID: AB_10973901Anti-SLC7A11Affinity BiosciencesCat# DF12509; RRID: AB_2845314Anti-HO-1GeneTexCat# GTX101147; RRID: AB_1950502Anti-FTH1HuaBioCat# ET-161078; RRID: AB_3069963Goat Anti-Mouse IgG H&L (Alexa Fluor® 488)AbcamCat# ab150113; RRID: AB_2576208Biological samplesMouse lung samplesThis paper–Mouse nasal samplesThis paper–Mouse tracheal samplesThis paper–Chemicals, peptides, and recombinant proteinsFTH1 (Homo sapiens) plasmidHanBioGene ID: 2495; transcript: NM_002032.3Cat# pHBPC001052Ferrostatin-1MedChemExpressCat# HY-100579EDTA decalcifying solutionSolarbioCat# E1171DMEMSolarbioCat# 12100RPMI-1640SolarbioCat# 31800Fetal Bovine Serum 1Lonsera Science SRLCat# S711-001Fetal Bovine Serum 2ExCell BioCat# FSP500LipoFiter 3.0HanBioCat# HB-LF3-1000Critical commercial assaysMouse IL6 ELISA KitMabtechCat# 3361-2HMouse TNF-α ELISA KitMabtechCat# 3511-2HTotal SOD activity Assay KitBeyotime BiotechnologyCat# S0101SLipid Peroxidation MDA Assay KitBeyotime BiotechnologyCat# S0131SROS Assay KitBeyotime BiotechnologyCat# S0033SpHrodo™ Green zymosan fluorescent bioparticlesInvitrogenCat# P35365HiScript II Q RT SuperMixVazymeCat# R222-01ChamQ Universal SYBRVazymeCat# Q711-02Deposited dataRNA-seqGEO database[GSE155616](GSE155616)RNA-seqGEO database[GSE294723](GSE294723)RNA-seqThis paperNovogene Science and Technology Co., Ltd.Experimental models: Cell linesBEAS-2BATCCCat# CRL-3588MH-SATCCCat# CRL-2019Experimental models: Organisms/strainsMouse: C57BL/6JBeijing SiPeiFu BiotechnologySCXK(JING)2019-0010Software and algorithmsImageJNIHhttps://imagej.net/software/fiji/downloadsCytoscape 3.8NRNBhttps://cytoscape.org/GraphPad Prism 8GraphPad Softwarehttps://www.graphpad.com/

Experimental model and study participant details

Mice

Male C57BL/6J mice, aged 10–12 weeks, were purchased from SiPeiFu Biotechnology Co., Ltd. (Beijing, China). All animals were housed and maintained in SPF conditions with 22 ± 2 °C, 50–60% humidity, and a 12/12 h light/dark cycle. All animals were acclimatized to laboratory conditions for seven days before proceeding to the in vivo studies. All animal experimental protocols were approved by the Animal Care and Ethics Committee of Henan University of Chinese Medicine (Approval No. DWLL202103158).

Mice models and treatment

To compare the respiratory toxicity of PM2.5 via three different exposure routes in mice, twenty-four of C57BL/6J mice were randomly divided into four groups: control group (Ctrl), PM2.5-exposed groups following intranasal instillation (NE), intratracheal instillation (TE), and whole-body exposure (WE). Experimental details are as follows: (1) PM2.5 suspension instillation: The collected PM2.5 powders were dissolved in saline to achieve a dose of 37.5 mg/mL (equivalent to 350 μg/m^3^ of PM2.5 in the ambient environment), which were intratracheally or intranasally instilled after mice were anesthetized using isoflurane inhalation. The controls received saline instillation. (2) PM2.5 aerosol inhalation: PM2.5 aerosol was obtained by an aerosol dispenser linked to an air compressor (HuiRongHe Technology Co., Ltd., Beijing, China). Mice in the whole-body exposure group were placed into a chamber for exposure to PM2.5 aerosol at a concentration of 350 μg/m^3^ for 4 h/day. The PM2.5 concentration within the chamber was continuously monitored using a PM2.5 concentration monitor (TSI instrument, MN, USA). The controls were exposed to clean filtered air. The PM2.5 exposure protocol was conducted once every two days for one week.

To explore the effect of Fer-1 on PM2.5-induced lung injury, eighteen mice were divided into three groups: control group (Ctrl), PM2.5 group, and Fer-1 group. Fer-1 was solubilized in DMSO and administered intraperitoneally (2 mg/kg) daily for 3 consecutive days before PM2.5 exposure, while Ctrl group and PM2.5 group were administered with DMSO solution. All groups were intratracheally instilled with PM2.5 suspension as described above, except Ctrl group. All groups of mice were sacrificed within 24 h following the final exposure, and the serum, bronchoalveolar lavage fluid (BALF), and lung tissues were collected for later experimental detection.

Cell lines

Human epithelial BEAS-2B cells (ATCC, CRL-3588) were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS, Lonsera Science SRL, Uruguay). Mouse alveolar macrophage MH-S cells (ATCC, CRL-2019) were maintained in RPMI-1640 medium supplemented with 10% FBS (ExCell Bio, Suzhou, China). All cell cultures were incubated in a humidified atmosphere containing 5% CO2 at 37 °C. All cell lines were tested for mycoplasma contamination using the Hoechst staining and were confirmed negative.

PM2.5 preparation

PM2.5 samples were obtained in December 2023 using a high-volume PM2.5 sampler (HY-1000 BL) operating at a flow rate of 1050 L/min. The particles were captured on a fiberglass filter membrane positioned on a building rooftop, 14 m above ground level, in Shijiazhuang. This location is in proximity to heavily trafficked roads and surrounded by residential areas. The collected filters were sectioned into smaller pieces and sonicated in sterilized water for 30 min at room temperature. Subsequently, PM2.5 suspensions were prepared by using eight layers of filter gauze, followed by complete freezing and lyophilization using a vacuum freeze-dryer. The resulting PM2.5 powders were then collected and stored at − 80 °C. Characteristics and chemical components, including polyaromatic hydrocarbons (PAHs), metals, and soluble ions of PM2.5 samples were described in Table S1.

Method details

Histopathology analysis

Tracheal and left lung segments were fixed immediately in a 4% (wt/vol) paraformaldehyde solution, followed by dehydration, paraffin embedding, and sectioning into 4 μm thick sections. For nasal histology, the nasal specimens were decalcified in an EDTA decalcifying solution (Solarbio, Beijing, China) for 72 h before paraffin embedding. The landmarks for sectioning were located immediately posterior to the incisors, at the level of the incisive papilla, and through the midpoint of the second molar tooth. All sections were subsequently deparaffinized using xylene and stained with hematoxylin and eosin (H&E) and periodic acid-Schiff (PAS) for histopathology and goblet cell metaplasia analysis. Images were captured under a light microscope (LEICA-DM6000B) or Pannoramic Viewer (3D HISTECH).

Oxidative stress and inflammatory cytokines measurements

The MDA and SOD levels in lung tissues and BEAS-2B were quantified by commercial MDA and SOD detection kits (Beyotime Biotechnology, China). Inflammatory cytokines in serum and BALF from mice and MH-S cells were quantified by the ELISA kit for mouse IL-6, and TNF-α (Mabtech, Sweden).

Pulmonary function

Pulmonary function was evaluated using a non-invasive whole-body plethysmograph (Buxco Inc., Wilmington, NC, USA) at two time points: within 1 h and 24 h following PM2.5 exposure. In brief, mice were anesthetized with 1.25% tribromoethyl aicolaol (125 mg/kg, ip) and placed in a whole-body volume tracking chamber. Under normal respiration, various parameters were recorded, including breathing frequency (f), minute ventilation volume (MV), tidal volume (TV), expiratory time (Te), pause-enhanced bronchial constriction (Penh), expiratory flow rate at 50% of TV (EF50), end-expiratory pause (EEP), peak expiratory flow (PEF), and peak inspiratory flow (PIF).

Immunofluorescence analysis

After dewaxing and dehydration, the 4 μm lung tissue sections were blocked by 5% normal goat serum for 1 h and then incubated with the primary antibodies against CD68 (Santa Cruz, California, USA), ZO-1, Occludin, E-cadherin (Proteintech, Wuhan, China), MUC5AC, MUC5B (Cloud-Clone, Wuhan, China), and CLCA1 (Abcam, Cambridge, MA, USA) (1:200) at 4 °C overnight. Subsequently, these slides were incubated with a secondary antibody (1:200) (Abcam, Cambridge, MA, USA) for 2 h at room temperature. Cell nuclei were counterstained with DAPI. Images were captured under a fluorescence microscope (Carl Zeiss, Oberkochen, Germany) and analyzed using ImageJ software.

BALF analysis

Following the induction of anesthesia, the left lung was ligated, and the upper portion of the trachea was cannulated. The right lung was then gently lavaged three times using phosphate-buffered saline through tracheal intubation. The recovered lavage fluid was centrifuged at 500 g for 10 min at 4 °C, and the supernatant was collected for inflammatory cytokines measurement. The cell pellets were smeared onto slides and stained with hematoxylin and eosin (H&E) for differential cell analysis by a LEICA-DM6000B microscope.

Cell treatment

To explore the cytotoxic effects of PM2.5, cells were treated with PM2.5 suspension at varying concentrations (10, 20, 50, 100 μg/cm^2^) over different time intervals (1, 3, 6, 12 h). To investigate the effect of Fer-1, all cells were pretreated with Fer-1 (5 μM) for 6 h before PM2.5 exposure (50 μg/cm^2^, 6 h).

Cell transfection

BEAS-2B and MH-S cells were cultured on 6-well plates and transiently transfected with an overexpression plasmid targeting FTH1 or the negative control (vector), following the manufacturer’s instructions of LipoFiter 3.0 (HanBio Co., Ltd, China). The FTH1 plasmid and negative control were synthesized by HanBio Co., Ltd. (Shanghai, China).

Western blot

Lung tissues and cells were lysed with cold RIPA Lysis Buffer plus PMSF. Protein concentrations were determined utilizing a BCA Protein Quantification Kit (Beyotime Biotechnology, P0010). The proteins were resolved on a 10% gel via SDS-PAGE and subsequently transferred onto PVDF membranes (Millipore, ISEQ00010). Following blocking with 5% nonfat milk, the membrane was incubated overnight at 4 °C with primary antibodies, including anti-ZO-1, anti-occludin, and anti-E-cadherin rabbit polyclonal antibodies (Proteintech, Wuhan, China). Then, the membrane was incubated with an HRP-conjugated secondary antibody for 1 h and developed by a SuperSignal West Dura chemiluminescent substrate kit (Thermo Scientific). Images were obtained using the Bio-Rad imaging system. The western blots were semi-quantified using ImageJ software to measure the intensities of the bands.

Quantitative RT-PCR

Total RNA was isolated from cells or lung tissue samples utilizing the QIAzol Lysis Reagent (Qiagen, Valencia, CA, USA) and subsequently reverse transcribed to cDNA using the HiScript II Q RT SuperMix (Vazyme, Nanjing, China). The qRT-PCR was performed on an Applied Biosystems 7500 platform (AB, CA, USA) using the ChamQ Universal SYBR (Vazyme, Nanjing, China). Specific primers were designed for each target mRNA (Table S2). The qPCR protocol comprised an initial denaturation phase at 95 °C for 30 s, followed by 40 cycles of denaturation at 95 °C for 10 s and annealing at 60 °C for 30 s. The levels of mRNA were normalized to the mRNA expression of GAPDH. Data analysis was performed using the comparative cycle threshold (ΔΔCt) method.

Transcriptomics analysis

Gene expression profiles related to PM2.5 exposure in BEAS-2B cells and macrophages were obtained from the GEO database. Inclusion criteria were: (1) mRNA expression profiles, and (2) PM2.5-induced BEAS-2B cells or macrophages. Consequently, datasets of [GSE155616](GSE155616) (BEAS-2B) and [GSE294723](GSE294723) (macrophages) were selected for analysis. For the investigation of PM2.5-induced lung injury in mice, RNA-sequencing (RNA-seq) was performed by Novogene Science and Technology Co., Ltd. (Beijing, China). Differentially expressed genes (DEGs) were identified based on the criteria of log2FC ≥ 1 and p < 0.05. Variations in mRNA expression levels were depicted in a volcano plot, and further analyses, including KEGG and GO enrichment analysis, were conducted using the DAVID platform with sample clustering, and data visualization were performed using the online platform http://www.bioinformatics.com.cn. In addition, a protein-protein interaction (PPI) network of the DEGs was constructed using the STRING database and visualized with Cytoscape 3.8 software.

Intracellular ROS production assessment

Reactive oxygen species (ROS) production was assessed using flow cytometry. Upon treatment with varying doses of PM2.5 in the 6-well plate over different time intervals, BEAS-2B cells were trypsinized and resuspended in fresh medium containing dichloro-dihydro-fluorescein diacetate (DCFH-DA, Solarbio, Beijing, China) for 30 min at 37 °C in the dark, followed by PBS washing and flow cytometry analysis (FACSCalibur, BD Biosciences, Carlsbad, CA, USA).

Phagocytic function analysis of MH-S

After overnight seeding in a laser confocal dish, MH-S cells were exposed to 20 μg/cm^2^ of PM2.5 for varying durations (0.5, 2, 6 h). Thereafter, the cells were incubated with pHrodo Green zymosan fluorescent bioparticles (Invitrogen, P35365) for 2 h, fixed with 4% paraformaldehyde for 10 min, and then sealed using an anti-fluorescence quenching tablet seal containing DAPI. The average number of bioparticles per cell and the percentage of cells containing bioparticles were quantified utilizing a laser scanning confocal microscope (Carl Zeiss, Oberkochen, Germany).

Quantification and statistical analysis

All data are shown as the mean ± SEM, and all experiments were repeated at least three times. Statistical analysis was conducted using GraphPad Prism 8 (GraphPad Software). Significant differences were assessed by one-way ANOVA followed by a Dunnett multiple comparisons test. A value of p less than 0.05 was considered statistically significant.